Welcome to the GitHub repository for CS-E4740 - Federated Learning, a master-level course offered every spring at Aalto University. This course introduces the foundations and applications of Federated Learning (FL)—a privacy-preserving and decentralized approach to training machine learning models on distributed data.
📘 Lecture notes are published as a Springer textbook:
Alexander Jung, Federated Learning: From Theory to Practice (Springer, 2025), arxiv preprint
- Formulate federated learning tasks as distributed optimization problems
- Design scalable and privacy-aware FL algorithms
- Understand the role of non-IID data, secure aggregation, and trustworthy AI
- Apply FL to real-world applications like weather prediction, healthcare, and recommendation systems
- ✅ Lecture Slides (based on the Springer textbook)
- 📓 Jupyter Notebooks and Python demos
- 🧪 Assignments and exercises
- 🧵 Real-world datasets for hands-on projects
- 📚 Additional readings on topics like differential privacy, robustness, and personalization
Enroll via Sisu. Contact your study coordinator for official registration.
Anyone can follow the course as open educational content. Subscribe to the course mailing list for updates.
📅 TBA
- 📙 Machine Learning: The Basics – Introductory ML textbook by Alexander Jung
- 🌐 My Personal Site
- 📺 YouTube Lectures
- 🌟 Star the repo to stay updated
- 🐛 Open issues for feedback or suggestions
- 🧠 Want to help? Fork the repo and suggest improvements or new examples
federated-learningdistributed-learningprivacy-preserving-ml
non-IIDsecure-aggregationoptimizationtrustworthy-ai
springer-textbookopen-coursewaredecentralized-ai
All content is released under the MIT License unless otherwise specified. Lecture slides and textbook excerpts follow publisher usage policy.
© 2025 Alexander Jung – Aalto University, Department of Computer Science